Cross-linking Effects on Performance Metrics of Phenazine-Based Polymer Cathodes

Article abstract of DOI:10.1002/cssc.201903243

Developing cathodes that can support high charge–discharge rates would improve the power density of lithium-ion batteries. Herein, the development of high-power cathodes without sacrificing energy density is reported. N,N′-diphenylphenazine was identified as a promising charge-storage center by electrochemical studies due to its reversible, fast electron transfer at high potentials. By incorporating the phenazine redox units in a cross-linked network, a high-capacity (223 mA h g^?1), high-voltage (3.45 V vs. Li/Li⁺) cathode material was achieved. Optimized cross-linked materials are able to deliver reversible capacities as high as 220 mA h g^?1 at 120 C with minimal degradation over 1000 cycles. The work presented herein highlights the fast ionic transport and rate capabilities of amorphous organic materials and demonstrates their potential as materials with high energy and power density for next-generation electrical energy-storage technologies.

Full text of DOI:10.1002/cssc.201903243

Accepted Article
Title: Cross-linking Effects on Performance Metrics of Phenazine-
Based Polymer Cathodes
Authors: Cara N. Gannett, Brian M. Peterson, Luxi Shen, Jeesoo Seok,
Brett P. Fors, and Héctor Abruña
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201903243
Link to VoR: http://dx.doi.org/10.1002/cssc.201903243
A Journal of
www.chemsuschem.org

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
Cross-linking Effects on Performance Metrics of Phenazine-
Based Polymer Cathodes
Cara N. Gannett †,[a] Brian M. Peterson†, [a] Luxi Shen, [a] Jeesoo Seok, [a] Brett P. Fors,* [a] and Héctor
D. Abruña*[a]
ion transport dynamics to improve the power density of these
devices.
Abstract: Developing cathodes that can support high charge-
discharge rates would improve the power density of lithium ion
batteries. Herein, we report the development of high power cathodes,
without sacrificing energy density. N,N’-Diphenyl phenazine was
identified as a promising charge storage center by electrochemical
studies due to its reversible, fast electron transfer at high potentials.
By incorporating the phenazine redox units in a cross-linked network,
a high capacity (223 mAh/g), high voltage (3.45 V vs Li/Li+) cathode
material was achieved. Optimized cross-linked materials are able to
deliver reversible capacities as high as 220 mAh g–1 at 120 C with
minimal degradation over 1000 cycles. The work presented here
highlights the fast ionic transport and rate capabilities of amorphous,
organic materials and demonstrates their potential as high energy and
power density materials for next generation electrical energy storage
technologies.
During charge and discharge, crystalline inorganic cathodes
undergo kinetically slow (de)intercalation events which limit the
rate of ion diffusion. In amorphous organic materials, charge
balancing counter ion transport can be dramatically enhanced
due to the relatively weak intermolecular forces within the
materials.[4,5] This enables amorphous, organic materials to
accommodate fast ion transport, resulting in faster charging and
discharging rates. However, designing amorphous, organic
materials that can deliver high power density without sacrificing
operating voltage, capacity, or stability has proven challenging.[6-
8]
Cross-linking in binders, electrolytes, and electrode materials,
provides greater resistance to dissolution, improved stability, and
supports an amorphous nature to increase ion conductivity.[9-13]
We hypothesized that the morphology of a redox active polymer
could be optimized by strategically tuning the cross-linking density.
Cross-linking can disrupt ordering and restrict polymer mobility,
ensuring an open, amorphous structure even in a swelled state.
Recently, we reported on a cathode material which utilized 3,7-
diamino-substituted phenothiazines as charge storage
elements.[14] In that study, cross-linking of a linear energy storage
polymer lead to improved stability upon cycling, higher discharge
capacities, and improved rate performance. Unfortunately, the
material exhibited poor capacity retention upon cycling, a likely
consequence of polymer degradation. In an effort to better
understand the effects of cross-linking, a polymer containing more
stable redox centers was necessary. Substituted phenazine
structures, specifically N,N’-diphenylphenazine (1), were
identified as a promising class of charge storage motifs through
electrochemical screening/analysis of phenazine-based small
molecule analogues.
Introduction
Electrification of transportation is expected to play a critical
role in curbing carbon emissions and utilizing renewable energy.
Recently, the US Department of Energy has set an initial target of
15 minute battery charging times to advance the market
penetration of electric vehicles.[1] Currently, capacitors are the
only electrical energy storage systems capable of stably
delivering and sustaining these rates, due to their fast rates of
charge storage.[2] However, capacitors lack faradaic reactions
resulting in low energy densities when compared to batteries,
compromising the range of electric vehicles. Current battery
materials have seen significant improvements in energy density
and lifetime over the past decade. However, the diffusive nature
of the energy storage processes continues to limit their rate
performance.[3] Thus, it is necessary to increase the charge and
In recent years, phenazine has been identified as a
promising charge storage unit in cathodes, capacitors, and as
redox flow catholyte.[15-24] Recent work on phenazine-based p-
type energy materials has been led by Zhang and coworkers, who
became interested in 1 for its ability to improve charge
delocalization about the dihydrophenazine unit.[23,24] In
a
collaborative effort with Zhou, they demonstrated the utility of a
crystalline phenazine-based bipolar molecule in a dual-ion
symmetric battery, with full cell capacities as high as 57 mAh g–1.
Additionally, they demonstrated the potential of a linear polymer
containing phenazine units as a high voltage cathode material,
delivering 65 mAh g–1 at a discharge rate of 5 C. However, in
neither accounts, were high C-rates reported.
Through ternary copolymerizations, we have synthesized
cross-linked polymers based around substituted phenazines and
studied the effects of cross-linking on their electrochemical
performance as amorphous cathode materials. These cross-
linked phenazine-based polymers delivered energy densities
competitive with commercial cathodes while providing
[a]
C. N. Gannett, † B. M. Peterson, † Dr. L. Shen, J. Seok, Prof. B. P.
Fors, Prof. H. D. Abruña
Chemistry and Chemical Biology
Cornell University
Ithaca, NY 14853-1301, United States
E-mail: brettfors@cornell.edu
E-mail: hda1@cornell.edu
[†]
These authors contributed equally to this work.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
dramatically enhanced power densities. By optimizing the cross-
linking density in these materials, we developed a cathode
material that can deliver a remarkable capacity of 220 mAh/g at
120 C with minimal loss in capacity.
+1 in 1, display ΔEpeak near the ideally reversible value (58 mV).
For the second redox couple from a ± 1 to ± 2 charge state, while
1 maintains a ΔEpeak of only 66 mV, the ΔEpeak of phenazine
increases to 216 mV, indicating an electrochemically irreversible
redox couple. Additionally, the ip,a/ip,c of the second redox couple
in phenazine decreases at increasing scan rates, again
corresponding to electrochemical irreversibility and potentially a
coupled chemical reaction (Figure S1). Together, this suggests
that materials based around 1, will display much higher operating
voltages and significantly more stable cycling in cathodes than
phenazine.
Results and Discussion
Electrochemical optimization of redox storage unit
The electrochemical properties of the redox active unit within a
material can dictate many of the performance metrics of the
resulting cathode, including operating voltage, cathode stability,
and rate capabilities. Thus, careful design and electrochemical
analysis of the redox unit, as a small molecule, can enable ideal
performance in the material. Alkylation and arylation of phenazine,
leads to improved kinetic and thermodynamic performance
metrics over unsubstituted phenazine. The modified N,N’-
disubstituted phenazine units cycle between a neutral and
dicationic state, belonging to a class known as p-type materials.
P-type materials typically possess higher operating voltages over
n-type materials (redox centers which undergo reduction and
become negatively charged) which is clearly advantageous for
cathode applications. Additionally, charge compensation within p-
type materials is established by anion transport from the
electrolyte. The smaller solvation shell of anions aids in providing
higher ionic diffusion and should enable higher charge/discharge
rates.[26]
Analysis of the cyclic voltammagrams of phenazine and
N,N’-diphenylphenazine (1) offers insight into the chemical and
electrochemical properties of the molecules. The redox processes
of 1 occur about 1500 mV more positive than those of phenazine
(Figure 1). This is ideal behavior for redox units in cathodes, as it
extends the operating voltage of the battery. The analysis of peak
to peak separation (ΔEpeak) and the ratio of the anodic peak
current to the cathodic peak current (ip,a/ip,c) of redox events sheds
insight into the chemical and electrochemical reversibility of the
redox centers (Table 1 and Table S1). The first redox couple,
transitioning from the neutral to a -1 state in phenazine and to a
Analysis of the voltammetric profiles of phenazine in the
presence and absence of trifluoroacetic acid (TFA) (to stabilize
the reduced state though binding of a proton), 1, and N,N’-
dimethylphenazine (2) at a Pt ultramicroelectrode was used to
determine the standard electron transfer rate constants (k0)
(Figure S3). In each case, the values of k0 indicated facile and
fast charge transfer kinetics (Table 1), and therefore should not
be rate limiting even at fast rates.[28] This small molecule study
ensures that rate performance of the reported materials in battery
applications will be limited by either electronic conductivity or ionic
diffusion through the material, and not the rate at which the redox
storage centers can be oxidized or reduced.
Table 1. Values of peak splitting, ΔEpeak, and standard electron transfer rate
constants, k°, for phenazine and small molecule derivatives.
ΔEpeak1
ΔEpeak2
Material
k°1 [cm/s]
2.6 × 10-2
1.4 × 10-2
k°2 [cm/s]
-
(0 → ±1)
(±1 → ±2)
Phenazine
Phenazine
+ 50 mM TFA
58 mV
216 mV
65 mV
107 mV
1.6 × 10-2
N,N’-dimethyl-
86 mV
65 mV
58 mV
66 mV
9.3 × 10-3
1.1 × 10-2
8.5 × 10-3
9.3 × 10-3
phenazine (2)
N,N’-diphenyl-
phenazine (1)
Design and characterization of storage materials
When designing a redox active polymer for electrochemical
energy storage (EES), any atoms that do not contribute to
faradaic processes should be limited. We hypothesized that the
core structure of 1 could be preserved and incorporated into a
partially conjugated polymer framework through the cross-
coupling of 5,10-dihydrophenazine (3) and dichlorobenzene (4).
Following a simple reduction of phenazine with sodium dithionite,
the newly freed secondary amines of 5,10-dihydrophenazine
provide excellent handles for Buchwald-Hartwig cross-coupling
reactions. Cross-coupling 3 with 4 yields poly(phenylene-5,10-
dihydrophenazine) (poly(Ph-PZ)), an insoluble powder which
precipitates from solution (Figure 2a). The ratio of 3 to 4 is given
in Table 2, entry 1. It can be seen by comparison of the infrared
spectra of poly(Ph-PZ) and 1, that the polymer shares a core
structure with
1
(Figure S4). As expected, the cyclic
voltammogram of the material, in a lithium metal half-cell, exhibits
two one-electron oxidations, similar to the CV profile of 1 (Figure
2b).
Cross-linked networks were synthesized to determine the
effect of cross-linking on the cathodes’ electrochemical properties.
We envisaged that tris(4-bromophenyl) amine (5) could be
copolymerized to cross-link the material (Table 2, entry 2) in order
to improve stability and disrupt ordering.[5,27,28] Copolymerization
of
3
with
5
results in
a
fully cross-linked material,
poly(triarylamine-5,10-dihydrophenazine) (poly(TAA-PZ)) which
exhibited much more resistance to dissolution in electrolyte
solution than poly(Ph-PZ). This was determined by UV-Vis
spectroscopy of filtered electrolyte solutions after suspending the
cathode material in the electrolyte solution for one week (Figure
Figure 1. The neutral and reduced states of phenazine and the neutral and
oxidized states of N,N’-diphenylphenazine (1). CV profiles at a glassy carbon
electrode of 1 mM solutions of phenazine (blue) and 1 (red) obtained in 0.1 M
TBAP in acetonitrile at 50 mV/s.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
Figure 2 (a.) Schematic of synthesis of poly(phenylene-5,10-dihydrophenazine) and cross-linked networks, (b.) CV of poly(Ph-PZ) in a Li metal coin cell with 1 M
LiPF6 in EC/DEC at 2 mV/s.
S5). Since any solubility of the active material will result in a loss
of capacity upon cycling, decreasing the solubility of the material
is critical to achieving near theoretical capacity and stable cycling
performance.
before 4 could be incorporated. This would leave a depleted
concentration of remaining in solution, resulting in
5
oligomerization between 3 and 4, as found in poly(Ph-PZ).
Therefore, while poly(Ph-PZ)-10 constitutes a homogenously
cross-linked material, it is more likely that poly(Ph-PZ)-50
behaves as a mixture of poly(Ph-PZ) and poly(TAA-PZ). The
morphology observed by SEM provides additional support for this
In our previous study of cross-linked poly(phenothiazine-
dimethylphenylenediamine), we observed that over a certain
threshold of cross-linking density, rate performance declined as a
result of lowered ionic transport through the material.[14] Therefore, hypothesis (Figure S6-8).
two molar ratios of 4 and 5 were examined to determine a cross-
linking density that would mitigate dissolution while retaining high
ion diffusion (Table 2, entries 3 and 4). The molar ratio in poly(Ph-
PZ)-10 lead to a decrease in solubility in the electrolyte solution
in comparison to poly(Ph-PZ) and poly(Ph-PZ)-50. It is important
to note that the degree of polymerization of the final polymer will
also play a significant role in determining the polymer solubility.
However, the cross-linked nature of the materials precludes
determination of molecular weight by conventional methods.
The crystallinity of the polymers was investigated by X-ray
diffraction (XRD). Diffractograms of the polymers revealed their
largely amorphous character (Figure S9). At 2θ values between
10°-25°, four small, broad peaks were observed in poly(Ph-PZ)
and poly(Ph-PZ)-10, indicating some ordering present in the
polymer. As expected, this ordering was lost as the amount of 5
was increased in poly(Ph-PZ)-50 and poly(TAA-PZ). However, no
Tm or Tc was observed up to 200 °C by differential scanning
calorimetry (DSC) in any of the synthesized polymers (Figure
S10). Therefore, any ordering observed by XRD is likely due to
local interactions between aromatic rings and does not suggest
any long range ordering within the materials.
Table 2. Reagent equivalents used in Buchwald-Hartwig cross-coupling
step-growth polymerizations to obtain phenazine-based cathode materials.
Entry
Polymer
(4) (equiv)
(5) (equiv)
(6) (equiv)
Coin cell design and cathode conductivity
1
2
3
4
Poly(Ph-PZ)
1.0
1.0
1.0
1.0
1.0
0.0
0.0
0.67
0.067
0.33
Poly(TAA-PZ)
Poly(Ph-PZ)-10
Poly(Ph-PZ)-50
For high power applications, fast electron transfer and ion
transport kinetics are necessary. To singularize the diffusion
limited rate of our materials, we aimed to minimize the rate
dependence on electronic transport by making the composite
conductivity sufficiently high.[3] Cathodes of each polymer were
prepared as composites with active material, Super P, CMK-3,
and poly(vinylidene difluoride) (PVDF) in a 3:3:3:1 ratio. The
partially conjugated π system in these materials is expected to
enable fast electron transport through the material[31-33] and the
high carbon content should provide rapid charge transport
through the composite, such that the electronic transport should
not be limiting. When not limited electronically, the ability of the
materials to transport ions should be rate limiting during charge
and discharge.
Potentiostatic electrochemical impedance spectroscopy
(PEIS) and cyclic voltammetry were used to measure the
cathodes’ charge transfer resistance and the ion diffusion
coefficients, respectively (Figure S15 and S16). The fits for the
charge transfer resistance of each of the plots are shown in Table
3 (additional fitting parameters are presented in Table S2). As
expected, all four polymers exhibited low charge transfer
0.90
0.50
The incorporation of cross-linker into the material was
ensured by taking advantage of the relative rates of monomer
cross-coupling reactions in the step growth polymerization. Due
to the faster rate of cross-coupling of 4 with aryl bromides over
aryl chlorides, we hypothesized that 6 would be more rapidly
incorporated into the polymer chains than 5, ensuring cross-linked
copolymers.[29,30] When the equivalent of 5 to 6 is low, as in
poly(Ph-PZ)-10, we expect that the cross-linker will be fully
incorporated within the polymer chains, decreasing the polymers
solubility under battery conditions. This is supported by the
elemental analysis of poly(Ph-PZ)-10 which revealed that no aryl-
bromide end groups were present. By contrast, elemental
analysis of
poly(Ph-PZ)-50 revealed 0.54 percent bromine
content and 0.67 percent chlorine content. The presence of
bromine in poly(Ph-PZ)-50 and the propensity of poly(Ph-PZ) for
dissolution, suggest that the copolymer of 3 and 5 exceeded its
solubility in toluene and precipitated from the reaction solvent
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
Figure 3. (a.) Charge-discharge profiles of tenth cycle of poly(Ph-PZ), poly(TAA-PZ), poly(Ph-PZ)-10, and poly(Ph-PZ)-50 at 5 C. (b.) Discharge capacity of poly(Ph-
PZ), poly(TAA-PZ), poly(Ph-PZ)-10, and poly(Ph-PZ)-50 at 5 C for 500 cycles.
resistance, enabling fast electron transport. The charge transfer
resistance of poly(Ph-PZ), poly(Ph-PZ)-10, and poly(Ph-PZ)-50
never exceeded 80 Ω throughout charging, indicating fast electron
transfer regardless of charge state (Figure S17).
phenazine units in poly(TAA-PZ), an integral ratio of 2:3 of the
area under the peaks in the CV would be expected if either peak
was due to oxidation of the triaryl amine units. However, we
observe an approximately 1:1 ratio in the CV’s of all of the
polymers (Figure S19). In all cases, the experimentally measured
capacities were slightly higher than the theoretical capacities of
the materials due to the additional capacity contribution from the
carbon in the cathode composite (Figure S20).
Continued cycling of poly(Ph-PZ), poly(Ph-PZ)-10, poly(Ph-
PZ)-50, and poly(TAA-PZ) at a C-rate of 5, illustrates the
prolonged stability of each material (Figure 3b, charge capacities
and coulombic efficiencies are shown in Figures S21-S24). All
polymers exhibited a modest loss in capacity over the first 50
cycles. Thereafter, they retained at least 92% capacity on the
500th cycle. Notably, poly(Ph-PZ)-10 still delivered a capacity of
196 mAh g–1 on the 500th cycle. It is evident that optimizing cross-
linking density to minimize dissolution while minimally affecting
capacity is an effective strategy to design a high energy cathode
material. At an average discharge voltage of 3.45 V, the poly(Ph-
PZ)-10 exhibits one of the highest energy densities
(approximately 770 Wh/kg of active material) of all polymeric
cathode materials reported.
Cyclic voltammograms of each cathode were taken over a
range of scan rates. Applying the Randles-Sevcik equation, the
diffusion coefficients of the PF6– ions through the electrode were
determined. Even at fast scan rates, the response remained
diffusion limited, as evidenced by the linear plot of peak current
vs. the square root of scan rate. The obtained diffusion
coefficients of poly(Ph-PZ) and poly(Ph-PZ)-10 are of the order of
–
10-8 cm2/s (Table 3). The faster diffusion of PF6 , observed in
poly(Ph-PZ) and poly(Ph-PZ)-10, compared to diffusion through
poly(Ph-PZ)-50 and poly(TAA-PZ), is attributed to improved ion
transport in the less densely cross-linked materials. For all the
polymers presented here, the values for the diffusion coefficient
were orders of magnitude higher than those of commercial
inorganic oxide cathodes, enabling high rate capabilities while
maintaining
a
high utilization of the active material.[34-37]
Table 3. Charge transfer resistances and diffusion coefficients obtained
from PEIS data at 2.7 V vs. Li/Li+ and CV data, respectively, for poly(Ph-
PZ), poly(TAA-PZ), poly(Ph-PZ)-10, and poly(Ph-PZ)-50.
The performance of each polymer under even more
demanding discharge rates was further analyzed. Figure 4a plots
capacity retention, at various discharge rates, as a percentage of
the polymer’s initial capacity averaged over the first three cycles
at 1 C. When discharge rates exceeded 5 C, charging rates were
held constant at a C-rate of 5 to ensure a constant initial charge
state, isolating the effect of rate on the discharge capacity. It is
evident that all of the materials excel under demanding loads as
they maintain the majority of their capacity even when discharged
at 60 C. Interestingly, poly(Ph-PZ)-10 had the highest capacity
retention at high rates of discharge, followed by poly(Ph-PZ)-50.
All four of the polymers recovered over 90% of their capacity when
the discharge rate was returned to 1 C.
Due to its high capacity, stability, and superior rate
performance, poly(Ph-PZ)-10 was cycled at the extraordinary
discharge rate of 120 C (Figure 4b). The material delivered a
capacity of 220 mAh g–1, corresponding to a power density of over
27 kW kg-1 when normalized to the mass of the entire cathode
composite. With continued cycling, the material stabilized as seen
in the improved coulombic efficiency with cycling, and the delivery
of 198 mAh g-1 after 1000 cycles. Further tests were run, probing
the performance of poly(Ph-PZ)-10 when charging and
discharging at increasing rates (Figure 4c). The charging voltage
was increased to 4.5 V to ensure full charge at high C-rates. The
material still delivered 180 mAh g–1 when charged and discharged
at 120 C and showed excellent capacity recovery when the
charge/discharge rate is returned to 0.5 C. At 120 C, poly(Ph-PZ)-
Polymer
Rct (Ω)
54.0
Dpk 1 (cm2/s)
1.6 × 10-8
4.0 × 10-9
1.6 × 10-8
5.3 × 10-9
Dpk 2 (cm2/s)
1.4 × 10-8
2.5 × 10-9
1.5 × 10-8
3.9 × 10-9
Poly(Ph-PZ)
Poly(TAA-PZ)
Poly(Ph-PZ)-10
Poly(Ph-PZ)-50
100.7
57.3
49.1
Material performance in lithium half-cells
The cathodes were cycled in lithium half-cells between 1.5 and
4.3 V at 5 C (Figure 3a), corresponding to a charge/discharge time
of 12 minutes; faster than the US Department of Energy target
charging rate. Poly(Ph-PZ)-10 exhibited the highest discharge
capacity of 223 mAh g–1, while the capacity of poly(Ph-PZ) upon
discharge was 209 mAh g–1. This improvement in capacity in
poly(Ph-PZ)-10, despite its lower theoretical capacity (203 mAh
g–1), is attributed to the aforementioned decrease in solubility over
poly(Ph-PZ). Due to the larger percentage of the inactive cross-
linker, poly(Ph-PZ)-50 and poly(TAA-PZ) delivered capacities of
187 mAh g–1 and 177 mAh g–1, respectively.
Since the redox activity of the triaryl amine units in the
polymers falls outside the electrochemical stability window (4.5 V),
they are not able to contribute any charge storage (Figure S18).
Through analysis of the CV of poly(TAA-PZ), each redox peak in
the CV could be attributed to oxidations from the embedded
phenazine. Since there are two triaryl amine units for every three
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
Figure 4. (a.) Discharge capacity retention of poly(Ph-PZ), poly(Ph-PZ)-10, poly(Ph-PZ)-50, poly(TAA-PZ) when discharged at increasing indicated C-rates. (b.)
Cycling performance of poly(Ph-PZ)-10, discharged at 120 C, over 1000 cycles. (c.) Discharge capacities and coulombic efficiencies of poly(Ph-PZ)-10 charged
and discharged at the denoted C-rates. (d.) Charge-discharge profile of poly(Ph-PZ)-10 charged and discharged at the indicated C-rates.
10 still retains its two discrete plateaus, indicative of a faradaic
charge storage mechanism (Figure 4d). These high rate
experiments illustrate the ability of amorphous materials to
provide fast ion transport and enable high rates of
charge/discharge. By tuning the active mass ratio used in the
composite, power and energy density can be optimized for
specific applications. The energy density of a cathode composite
can be driven higher by increasing the active mass loading. Cells
with 60% active material in the composite were tested at varying
discharge rates. At low C-rates (0.5 C) poly(Ph-PZ)-10 was able
to deliever a capacity of 182 mAh g-1 (Figure S28). When the cell
was discharged at progressively higher C-rates (60 C), poly(Ph-
PZ)-10 delivered a capacity of 143 mAh g-1. Future work will focus
on further understanding ionic transport through amorphous
organic materials.
Performance in sodium half-cells
P-type cathode materials lend themselves to be used in
alternative ion batteries. Since anions are responsible for the
charge compensation of the oxidized phenazine units, materials
based around this redox active unit are largely unaffected by the
metal cations in solution. Therefore, the materials presented here
could be used in a range of batteries employing anodes such as
sodium, magnesium, and potassium without the performance of
the cathode being affected. This effect is demonstrated by
analyzing the CV profile of poly(Ph-PZ)-10 in a variety of
electrolyte salts, shown in Figure S29. The redox peaks of
poly(Ph-PZ)-10 retain their shape and magnitude regardless of
the nature of the cation in solution.
We sought to determine the eficacy of a poly(Ph-PZ)-10
cathode in a sodium half cell, as sodium ion batteries are touted
as being a future alternative EES system to lithium ion batteries
(Figure 5). Coin cells were assembled with a sodium metal anode
and a poly(Ph-PZ)-10 composite as the cathode with 1 M NaPF6
in EC/DEC as the electrolyte. Due to a large resistance in the
sodium cell, the coin cell was tested at a relatively low C-rate.
Cycling the cell at 0.5 C provided a steady discharge capacity of
over 160 mAh g–1 after 100 cycles, comparable to the capacity
obtained in a lithium cell. The large charging capacity in the first
cycle and the poor coulombic efficiency are attributed to the
incompatibility of carbonate-based electrolytes with sodium metal
deposition, leading to an increased resistance in the cell, as
observed by PEIS (Figure S30).[38,39] To obtain the same results
achieved in the lithium cell, each of the components of the sodium
coin cell would require further optimization. Still, these results
Figure 5. Cycling performance of poly(Ph-PZ)-10 in a sodium metal half-cell at
0.5 C. Inset: Charge-discharge profile of sodium metal half-cell.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
demonstrate the ability of poly(Ph-PZ)-10 to function in alternative
ion batteries.
0.1 M TBAP in MeCN unless otherwise specified. The potentials
in the CVs were corrected for iR drop using a ferrocene
reference at different scan rates.
Conclusions
Synthesis of 5,10-dihydrophenazine
The performance metrics of the phenazine-based materials
described here, as cathodes, are among the highest reported for
organic cathode materials. The careful design of a cross-linked
copolymer, poly(Ph-PZ)-10, resulted in improved capacity,
stability, and rate performance over the linear poly(Ph-PZ). At an
average operating voltage of 3.45 V vs. Li/Li+, poly(Ph-PZ)-10
delivers 223 mAh g–1 and 220 mAh g–1 at 5C and 120C,
respectively. The addition of the cross-linker was shown to enable
a higher active material ratio to be used in the cell while
maintaining a high percentage of the material’s capacity. The
family of materials studied here exhibited excellent ionic transport
kinetics that resulted in extraordinary capacity retention at very
high C-rates. These materials possess energy densities
competitive with current commercial lithium ion batteries and
power densities comparable to commercial capacitors, while
demonstrating the viability of organic cathodes in EES devices.
Specifically, this work demonstrates the impact that fast ionic
diffusion through amorphous materials has on rate capability.
Ethanol and DI water were degassed by sparging with nitrogen
for 30 min. 400 mL of water and 100 mL of ethanol were added
to a 1 L round bottom flask. Phenazine (4.00 g, 22.2 mmol, 1
equiv) and sodium dithionite (46.6 g, 268 mmol, 12 equiv) were
added and the solution was magnetically stirred under a nitrogen
atmosphere. The reaction was heated to reflux under nitrogen
and allowed to proceed for 4 hrs. The reaction was deemed
complete when no solid blue particulate remained. The reaction
was allowed to cool, quickly filtered, and washed with
deoxygenated water, yielding 3.57 g (88 % yield) of a light green
powder. The product was dried and stored under vacuum until
use.
General Procedure for Buchwald-Hartwig Cross-coupling
Polymerizations
The following method was implemented for the synthesis of
poly(Ph-PZ), poly(Ph-PZ)-10, poly(Ph-PZ)-50, and poly(TAA-PZ)
(the amount of reagents are detailed for each polymer in the
following sections). 5,10-dihydrophenazine, dichlorobenzene,
tris(4-bromophenyl)amine, sodium tertbutoxide, RuPhos ligand,
and RuPhos Pd G2 precatalyst were charged to a schlenk tube.
A nitrogen atmosphere was established and toluene was added.
The reaction was stirred at 110 °C for 16 hours. Upon
Experimental Section
General Synthetic Procedures
completion of the reaction time, the polymer was suspended in
dichloromethane and washed with water 3 times. After filtration,
the polymer was dried at 65 °C under vacuum.
General information regarding reagents can be found in the
Supporting information. N,N’-dimethyl-5,10-dihydrophenazine
and N,N’-diphenyl-5,10-dihydrophenazine were synthesized
employing previously reported methods.[40,41]
Synthesis of Poly(Ph-PZ)
The general procedure for Buchwald-Hartwig cross-coupling
polymerization was followed using 5,10-dihydrophenazine (273
mg, 1.50 mmol, 1 equiv), 1,4-dichlorobenzene (221 mg, 1.50
mmol, 1 equiv), RuPhos (14 mg, 0.03 mmol, 0.02 equiv),
RuPhos Pd G2 precatalyst (23 mg, 0.03 mmol, 0.02 equiv), and
NaOtBu (346 mg, 3.6 mmol, 2.4 equiv). The polymer (357 mg)
was obtained as a tan powder. IR (ATR, cm–1): 3033, 1604,
1503, 1476, 1455, 1329, 1260, 1093, 1061, 1015, 908, 817, 723,
620, 559. Elemental Anal. Found: C, 79.63; H, 4.45; Cl, 2.71.
Fabrication of Coin-Cells
Device measurements were conducted in CR 2032 coin cell
casings, which were assembled in an argon filled glove box with
oxygen levels below 1 ppm. The anode was lithium metal and
the working electrode was fabricated from a slurry of 30 wt%
active material (~0.6 mg/cm2), 30 wt% Super P carbon, 30 wt%
CMK-3, and 10 wt% PVDF as the binder in NMP. The high
loading cathode was fabricated with 60 wt% active material, 15
wt% Super P carbon, 15 wt% CMK-3, and 10 wt%
Synthesis of Poly(Ph-PZ)-10
poly(vinylidene fluoride) (PVDF) as binder. The slurries were
spread onto a carbon paper current collector using the doctor
blade method with one layer of scotch tape. The coated
electrode was dried for 2 h at 60 °C followed by 14 h at 110 °C
in a vacuum oven. A polypropylene separator (Celgard 2300) or
a glass microfiber filter (GF/A, Whatman) was used as the
separator between the two electrodes. The electrolyte solution
was 1:1 by volume of EC (ethylene carbonate) to DEC (diethyl
carbonate) with 1 M LiPF6 (Aldrich). In cells with the celgard
separator, 60 µL of electrolyte solution were used, while 80 µL
were used in the cells with the fiber glass separators. An Arbin
BT2000 battery tester was used to perform galvanostatic
charge-discharge experiments on the coin cells at 25 ° C.
The general procedure for Buchwald-Hartwig cross-coupling
polymerization was followed using 5,10-dihydrophenazine (273
mg, 1.50 mmol, 1 equiv), 1,4-dichlorobenzene (198.5 mg, 1.35
mmol, 0.9 equiv), tris(4-bromophenyl)amine (48.2 mg, 0.1 mmol,
0.067 equiv), RuPhos ligand (14 mg, 0.03 mmol, 0.02 equiv),
RuPhos Pd G2 precatalyst (23 mg, 0.03 mmol, 0.02 equiv), and
NaOtBu (346 mg, 3.6 mmol, 2.4 equiv). The polymer (394 mg)
was obtained as a brown powder. IR (ATR, cm–1): 3033, 1603,
1502, 1476, 1455, 1329, 1259, 1156, 1093, 1061, 1015, 920,
817, 723, 620, 559. Elemental Anal. Found: C, 80.50; H, 4.71;
Cl, 1.19.
Synthesis of Poly(Ph-PZ)-50
Procedure for cyclic voltammetric experiments
The general procedure for Buchwald-Hartwig cross-coupling
polymerization was followed using 5,10-dihydrophenazine (273
mg, 1.50 mmol, 1 equiv), 1,4-dichlorobenzene (110 mg, 0.75
mmol, 0.5 equiv), tris(4-bromophenyl)amine (241 mg, 0.5 mmol,
0.33 equiv), RuPhos ligand (14 mg, 0.03 mmol, 0.02 equiv),
RuPhos Pd G2 precatalyst (23 mg, 0.03 mmol, 0.02 equiv), and
NaOtBu (346 mg, 3.6 mmol, 2.4 equiv). The polymer (439 mg)
was obtained as a brown powder. IR (ATR, cm–1): 2079, 1605,
1499, 1479, 1455, 1333, 1280, 1259, 1156, 1093, 1061, 1015,
Benchtop cyclic voltammetry experiments were performed in a
three-compartment glass cell with medium porosity glass frits
separating the compartments. An Ag/Ag+ reference electrode
and a Pt wire counter electrode were used. All 3-mm glassy
carbon (GC) electrodes were purchased from CH Instruments
and used as working electrodes. Prior to the experiment,
electrodes were polished using diamond paste and MetaDi Fluid
(Buehler). All cyclic voltammetric measurements were done in
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
[22] G. Kwon, S. Lee, J. Hwang, H. S. Shim, B. Lee, M. H. Lee, Y. Ko,
S. K. Jung, K. Ku, J. Hong, K. Kang, Joule, 2018, 2, 1771.
[23] G. Dai, X. Wang, Y. Qian, Z. Niu, X. Zhu, J. Ye, Y. Zhao, X. Zhang,
Energy Storage Mater. 2019, 16, 236.
[24] Z. Niu, H. Wu, L. Liu, G. Dai, S. Xiong, Y. Zhao, X. Zhang, J. Mater.
Chem. A, 2019, 7, 10581.
820, 730, 617, 555. Elemental Anal. Found: C, 79.96; H, 4.69;
Br, 0.54; Cl, 0.67.
Synthesis of Poly(TAA-PZ):
[25] K. Xu, Chem. Reviews, 2004, 104, 4303.
[26] A. Z. Weber, M. M. Mench, J. P. Meyers, P. N. Ross, J. T. Gostick,
Q. Lui, J. Appl. Electrochem. 2011, 41, 1137.
[27] J. F. Le Nest, S. Callens, A. Gandini, M. Armand, Electrochim.
Acta, 1992, 37, 1585.
[28] O. Garcia-Calvo, N. Lago, S. Devaraj, M. Armand, Electrochim.
Acta, 2016, 220, 587.
[29] F. Barrios-Lnderos, B. P. Carrow, J. F. Hartwig, J. Am. Chem. Soc.
2009, 131, 8141.
[30] B. P. Fors, S. L. Buchwald, J. Am. Chem. Soc. 2009, 131, 12898.
[31] Peng, G.-H. Ning, J. Su, G. Zhong, W. Tang, B. Tian, C. Su, D. Yu,
L. Zu, J. Yang, Nat. Energy, 2017, 2, 17074.
The general procedure for Buchwald-Hartwig cross-coupling
polymerization was followed using 5,10-dihydrophenazine (273
mg, 1.50 mmol, 1 equiv), tris(4-bromophenyl)amine (482 mg, 1.0
mmol, 0.67 equiv), RuPhos ligand (14 mg, 0.03 mmol, 0.02
equiv), RuPhos Pd G2 precatalyst (23 mg, 0.03 mmol, 0.02
equiv), and NaOtBu (346 mg, 3.6 mmol, 2.4 equiv). The polymer
(414 mg) was obtained as a brown powder. IR (ATR, cm–1):
3028, 1606, 1497, 1479, 1456, 1310, 1282, 1256, 1158, 1099,
1060, 1013, 822, 817, 732, 616, 553. Elemental Anal. Found: C,
76.83; H, 4.59; Br, 1.49.
[32] Castillo-Martínez, J. Carretero-González, M. Armand, Angew.
Chem. Int. Ed. 2014, 53, 5341.
[33] G. P. Evans, in Advances in Electrochemical Science and
Engineering, (Eds: H. Gerischer and C. Tobias), VCH Publishers
Inc., 1990, pp. 3-8.
[34] X. Wang, H. Hao, J. Liu, T. Huang, A. Yu, Electrochim. Acta, 2011,
56, 4065.
[35] S. B. Tang, M. O. Lai, L. Lu, Mater. Chem. Phys. 2008, 111, 149.
[36] S. Yang, X. Wang, X. Yang, Y. Bai, Z. Liu, H. Shu, Q. Wei,
Electrochim. Acta, 2012, 66, 88.
[37] Y. Cui, X. Zhao, R. Guo, Electrochim. Acta, 2010, 55, 922.
[38] H. Kim, J. Hong, Y. U. Park, J. Kim, I. Hwang, K. Kang, Adv. Funct.
Mater. 2015, 25, 534.
[39] G. Yan, D. Alves-Dalla-Corte, W. Yin, N. Madern, G. Gachot, J.-M.
Tarascon, J. Electrochem. Soc. 2018, 165, A1222.
[40] A. Sugimoto, T. Kotani, J. Tsugimoto, S. Yoneda, J. Heterocycl.
Chem., 1989, 26, 435.
[41] J. C. Theriot, C.-H. Lim, H. Yang, M. D. Ryan, C. B. Musgrave, G.
M. Miyake, Science, 2016, 352, 1082.
Acknowledgements
This work was primarily funded by a grant to the Cornell Center
for Materials Research from the NSF MRSEC program (DMR-
1719875).
Keywords: electrochemistry • cross-coupling • nitrogen
heterocycle • organic cathode • phenazine
[1] D. Howell et. al., Enabling Fast Charging: A Technology Gap
Assessment, https://www.osti.gov/servlets/purl/1416167, accessed:
September, 2019.
[2] C. Schütter, S. Pohlmann, A. Balducci, Adv. Energy Mater., 2019, 9,
1900334.
[3] R. Tian, S. H. Park, P. J. King, G. Cunningham, J. Coelho, V.
Nicolosi, J. N. Coleman, Nat. Commun., 2019, 10, 1933.
[4] M. Yao, K. Kuratani, T. Kojima, N. Takeichi, H. Senoh, T.
Kiyobayashi, Sci. Rep. 2014, 4, 3650.
[5] M. Armand, Solid State Ion., 1983, 9, 745.
[6] D. R. Nevers, F. R. Brushett, D. R. Wheeler, Jour. of Power Sources,
2017, 352, 226.
[7] B. Haeupler, A. Wild, U. S. Schubert, Adv. Energy Mater., 2015, 5,
1402034.
[8] A. M. Bryan, L. M. Santino, Y. Lu, S. Acharya, J. M. D’Arcy, Chem.
Mater., 2016, 28, 5989.
[9] Y. Ma, J. Ma, G. Cui, Energy Storage Mater. 2018, 20, 146.
[10] S. J. Tan, X. X. Zeng, Q. Ma, X. W. Wu, Y. G. Guo, Electrochem.
Energy Rev. 2018, 1, 113.
[11] P. T. Dirlam, R. S. Glass, K. Char, J. Pyun, J. Polym. Sci. A, 2017,
55, 1635.
[12] S. Muench, A. Wild, C. Friebe, B. Hꢀupler, T. Janoschka, U. S.
Schubert, Chem. Rev., 2016, 116, 9438.
[13] J. Lopez, D. G, Mackanic, Y. Cui, Z. Bao, Nat. Rev. Mater., 2019, 4,
312.
[14] B. M. Peterson, D. Ren, L. Shen, Y.-C. M. Wu, B. Ulgut, G. W.
Coates, H. D. Abruña, B. P. Fors, ACS Appl. Energy Mater. 2018, 1,
3560.
[15] M. Lee, J. Hong, B. Lee, K. Ku, S. Lee, C. B. Park, K. Kang, Green
Chem. 2017, 19, 2980.
[16] S.-Y. Yang, Y.-J. Chen, G. Zhou, Z.-W. Fu, J. Electrochem. Soc.
2018, 165, A1422.
[17] J. Wang, K. Tee, Y. Lee, S. N Riduan, Y. Zhang, J. Mater. Chem. A
2018, 6, 2752.
[18] T. Matsunaga, T. Kubota, T. Sugimoto, M. Satoh, Chem. Lett. 2011,
40, 750.
[19] Hollas, X. Wei, V. Murugesan, Z. Nie, B. Li, D. Reed, J. Liu, V.
Sprenkle, W. Wang, Nat. Energy 2018, 3, 508.
[20] Peng, G.-H. Ning, J. Su, G. Zhong, W. Tang, B. Tian, C. Su, D. Yu,
L. Zu, J. Yang, Nat. Energy 2017, 2, 17074.
[21] J. Wang, C. S. Chen, Y. Zhang, ACS Sustain. Chem. Eng. 2017, 6,
1772.
This article is protected by copyright. All rights reserved.

10.1002/cssc.201903243
ChemSusChem
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Cross-linked Cathodes: A
phenazine based polymer,
poly(phenylene-5,10-
Cara N. Gannett, Brian M. Peterson,
Luxi Shen, Jeesoo Seok, Brett P.
Fors*, Héctor D. Abruña*
dihydrophenazine), is cross-linked
with tris(4-bromophenyl) amine to
form a cross-linked network with
improved stability and rate
capabilities.
Page No. – Page No.
Cross-linking Effects of
Performance Metrics of Phenazine-
Based Polymer Cathodes
This article is protected by copyright. All rights reserved.